[Intel-gfx] ✗ Fi.CI.BAT: failure for How far have we got in converting the world to igt_dummyload?

[Chris Wilson]

Quoting Patchwork (2019-01-24 08:13:50)
> == Series Details ==
> 
> Series: How far have we got in converting the world to igt_dummyload?
> URL   : https://patchwork.freedesktop.org/series/55670/
> State : failure
> 
> == Summary ==
> 
> CI Bug Log - changes from CI_DRM_5472 -> IGTPW_2290
> ====================================================
> 
> Summary
> -------
> 
>   **FAILURE**
> 
>   Serious unknown changes coming with IGTPW_2290 absolutely need to be
>   verified manually.
>   
>   If you think the reported changes have nothing to do with the changes
>   introduced in IGTPW_2290, please notify your bug team to allow them
>   to document this new failure mode, which will reduce false positives in CI.
> 
>   External URL: https://patchwork.freedesktop.org/api/1.0/series/55670/revisions/1/mbox/
> 
> Possible new issues
> -------------------
> 
>   Here are the unknown changes that may have been introduced in IGTPW_2290:
> 
> ### IGT changes ###
> 
> #### Possible regressions ####
> 
>   * igt at gem_exec_reloc@basic-cpu-active:
>     - fi-skl-6700hq:      PASS -> INCOMPLETE +16
> 
>   * igt at gem_exec_reloc@basic-cpu-gtt-active:
>     - fi-whl-u:           PASS -> INCOMPLETE +17
>     - fi-skl-6770hq:      PASS -> INCOMPLETE +16
>     - fi-bsw-n3050:       PASS -> INCOMPLETE +14
>     - fi-kbl-7560u:       PASS -> INCOMPLETE +17
>     - fi-skl-iommu:       PASS -> INCOMPLETE +16
>     - fi-skl-6260u:       PASS -> INCOMPLETE +16
> 
>   * igt at gem_exec_reloc@basic-cpu-read-active:
>     - fi-ilk-650:         PASS -> INCOMPLETE +16
>     - fi-ivb-3520m:       PASS -> INCOMPLETE +17
> 
>   * igt at gem_exec_reloc@basic-gtt-cpu-active:
>     - fi-kbl-8809g:       PASS -> INCOMPLETE +12
> 
>   * igt at gem_exec_reloc@basic-gtt-read-active:
>     - fi-kbl-x1275:       PASS -> INCOMPLETE +17
>     - fi-kbl-r:           PASS -> INCOMPLETE +17
>     - fi-gdg-551:         PASS -> INCOMPLETE +14
> 
>   * igt at gem_exec_reloc@basic-write-cpu-active:
>     - fi-hsw-4770:        PASS -> INCOMPLETE +16
>     - fi-skl-guc:         PASS -> INCOMPLETE +16
>     - fi-kbl-7500u:       PASS -> INCOMPLETE +17
> 
>   * igt at gem_wait@basic-busy-all:
>     - fi-hsw-4770r:       PASS -> INCOMPLETE +16
>     - fi-kbl-guc:         PASS -> INCOMPLETE +12
>     - fi-blb-e6850:       PASS -> INCOMPLETE +15
>     - fi-bwr-2160:        PASS -> INCOMPLETE +14
> 
>   * igt at kms_busy@basic-flip-a:
>     - fi-kbl-7567u:       PASS -> INCOMPLETE +15
>     - fi-skl-6700k2:      PASS -> INCOMPLETE +16
> 
>   * igt at kms_busy@basic-flip-b:
>     - fi-bdw-5557u:       PASS -> INCOMPLETE +17
>     - fi-cfl-guc:         PASS -> INCOMPLETE +17
> 
>   * igt at kms_busy@basic-flip-c:
>     - fi-cfl-8109u:       PASS -> INCOMPLETE +16
> 
>   * igt at kms_cursor_legacy@basic-busy-flip-before-cursor-legacy:
>     - fi-pnv-d510:        PASS -> INCOMPLETE +15
>     - fi-hsw-peppy:       PASS -> INCOMPLETE +16
>     - fi-cfl-8700k:       PASS -> INCOMPLETE +17

We really should handle igt_skip from inside igt_fork better.

The should have been recorded as CRASH for starters.

Any way at that count, we have a grand total of 7 tests in BAT using the
spinner. I thought we had converted much more.
-Chris